Somebody has told you to map your process.
Maybe a consultant. Maybe an article like this one. Maybe the software you are being sold needs one before it will do anything useful. So you opened a blank page, drew a box, wrote "enquiry comes in" inside it, then sat there.
Then what?
That is where most process mapping dies. Not through laziness. The instruction gets handed over as though the finished thing is obvious. It is not. Nobody tells you what you are aiming at, how much detail is enough, or how you would know when you had finished.
So this is the missing half. What the finished article looks like, how to draw one in an afternoon, then how to read it afterwards to find the step that is quietly costing you the most.
If you have not run the bottleneck audit yet, start there. That one finds the hours you cannot see. This one shows you where in the work they are going.
A process map is not a flowchart
Clear this out of the way first, because it stops more people than anything else.
You have seen flowcharts. Diamonds for decisions, rectangles for actions, arrows in every direction, usually drawn in software with a lot of colour. They look professional. For what you are trying to do they are mostly useless.
A flowchart shows the logic of a process. You already know the logic. You run the business. What you do not know is where the time goes. A flowchart is silent on that.
A process map worth an afternoon is closer to a list than a picture. One line per step, in the order the work really happens, with four things written against each one.
Who does it. How long it takes them. What they need before they can start. How long it then sits there waiting.
That last column is the entire point. It is also the one nearly every map leaves out.
Touch time is not elapsed time
Here is the distinction that makes this exercise worth doing at all.
Touch time is how long the work takes while somebody is actually doing it.
Elapsed time is how long it takes from the customer's side, first contact to finished, including every minute it sat in a pile.
Ask an owner how long it takes to get a quote out and the answer is usually something like forty minutes. That answer is honest. Somebody does sit down for forty minutes and produce a quote.
Now map the same thing properly.
One quote, mapped
This is a composite drawn from several firms, but the shape is real and I have seen it more than once.
Seven steps. For each one, who does it, how long the work takes them, then how long the job sits still before the next step starts.
The enquiry arrives by email. Office, two minutes to read it and log it. It had already sat four hours before anybody opened it.
Call back, book a site visit. Office, ten minutes. Then six days until the first free slot in the diary.
Visit the site. Owner, forty five minutes. Then two days before the notes get typed up.
Type up the notes, work out the materials. Owner, thirty minutes. Then one day waiting on a price back from a supplier.
Write the quote. Office, twenty minutes. Then one day sitting in the owner's tray.
Owner reads it, signs it off. Owner, five minutes. No wait on this one.
Send it. Office, two minutes. Done.
Touch time, added up, is one hour and fifty four minutes.
Elapsed time is a little over ten days.
Read those two numbers next to each other, because the gap between them is the whole article. Under two hours of work is spread across ten days. For roughly ninety nine per cent of that time, nobody is touching the job at all. It is sitting still.
Your customer did not experience two hours. They experienced ten days of silence. The competitor who got a quote to them on Wednesday won the job. It was not because they worked harder or priced better. They just had fewer places for it to stop.
The expensive step is almost never the long one
Now you can read the map. Three things to mark on it, in this order.
The waits. Go down the waits and find the biggest number. In the example it is the six day gap waiting for a site visit slot, which on its own is more than half the elapsed time. That is the step to attack first. Notice that nobody in the business is doing anything wrong during it. There is no lazy employee to find. The delay is structural.
The handoffs. Every time the work crosses from one person to another, draw a mark. The example has four crossings in seven steps. Handoffs are where things stop, where context gets lost, where somebody has to ask a question before they can start. A process with four handoffs will almost always beat the same process with nine. When you are looking for something to change, a handoff you can delete is usually worth more than a task you can speed up.
The loops. Anywhere work comes back to a step it has already been through, draw an arrow backwards. That is rework. Quote goes to the owner, owner spots a missing item, quote goes back. Each loop costs you the touch time twice and adds a fresh wait on the end. Loops are the most under-reported thing on any map, because the people doing the work think of them as normal rather than as a fault.
Here is the rule that surprises people. The step with the longest touch time is rarely the one worth fixing. In the example it is the site visit at forty five minutes. The site visit is the job. You do not want to speed that up. The four hours before anyone opens the email is where the money is. It is also the cheapest thing on the sheet to change.
How to draw one, in an afternoon
Paper. Not software. I mean that.
Pick one process that a customer would recognise. Enquiry to quote. Order to delivery. Job finished to invoice paid. Something with a start and an end that somebody outside the business would understand. Do not try to map the whole company. That is how this turns into a three week project nobody finishes.
Get the people who actually do it in a room. Not their manager. Not you on your own with a coffee. The person who opens the email and the person who types the quote will disagree about what happens between them. That disagreement is the most valuable thing you will hear all afternoon.
One step per sticky note, in a line along a wall. The only question you ask is "then what happens?" Keep asking it until you reach the end. Do not tidy as you go.
Then go back along the line and add the four columns. Who, touch time, what they need first, then how long it waits. Rough is fine. "About a day, sometimes three" is a perfectly good entry and it is far more useful than a made-up precise figure.
Ask about the waiting out loud, every single time. "And then how long does it usually sit there?" People will not volunteer this. In their heads the job moves from one task to the next with nothing in between, because the waiting is not their doing. You have to ask for it step by step or your map will show two hours of work and nothing else.
An afternoon is enough. If it takes three days you have picked too much.
The five ways this goes wrong
Mapping how it should work. The map you want is the one that shows Tuesday as Tuesday actually is, including the workaround somebody invented in 2023 because the proper system was too slow. If the finished map looks tidy, it is wrong. Say so at the start: we are drawing what happens, not what is supposed to happen. Nobody is in trouble for what goes on the wall.
Mapping too wide. One process. If yours has more than about fifteen steps you have probably strung three processes together.
Reaching for software too early. Drawing software makes an ugly process look neat, which is the opposite of helpful. It also quietly encourages you to spend the afternoon on layout instead of on the question. Paper first, always. If you want it tidy afterwards, tidy it afterwards.
Doing it alone. You will map the version in your head. Yours is the least accurate version in the building, because you have not done most of these steps for years.
Stopping at the map. A map on a wall changes nothing. It is a diagnostic, not a treatment. The point is the next paragraph.
What finished looks like
You know you have finished when the sheet passes three tests.
It fits on one page and somebody who does not work for you can follow it.
It has numbers on it, not adjectives. "Slow" is not a finding. "Six days" is.
It shows the waiting as clearly as it shows the work, so the touch total and the elapsed total sit next to each other where you can see both.
That is the finished article. Not beautiful. One sheet of paper with a line of steps, four columns, some marks where the work crosses people, a few backwards arrows, then two totals at the bottom that do not match.
Then pick one thing
Take the biggest wait on the sheet and ask one question about it. Why does it wait there?
The answer is usually one of four. Somebody has to approve it. Somebody has to be told it is ready. The information needed is somewhere else. Or the person it goes to next is busy with something that could have been done at another time.
The first three of those are often fixable without buying anything at all. A standing rule that quotes under a certain value do not need signing off. A shared list instead of an email. Supplier prices held in one place rather than asked for each time. I would rather you found one of those and fixed it yourself than spent a penny with anybody.
The fourth one is where automation can genuinely earn its keep, because a task that is short and repetitive and rules-based is exactly the sort of thing a machine handles well. That is also the honest limit of it. If your bottleneck is a person waiting on a decision only you can make, no software on the market will touch it. Anybody who tells you otherwise is selling.
Where to start this week
Take a pad into the office tomorrow and write down the seven steps of one job, end to end. Do not worry about the columns yet. Just the steps, in order, as they really run.
Then go and ask one person who does one of those steps how long it sits with them before they get to it. The answer will be longer than you expect.
That is the whole exercise in miniature. It takes about twenty minutes.
If you get through the full afternoon and end up looking at a sheet you are not sure how to act on, that is what we do. The first conversation is free and there is no pitch attached to it. Bring the sheet. If the honest answer is that you have already found your own fix and do not need us, you will hear that.
If it does make sense to go further, the automation side of what we do starts in exactly this place, with the admin nobody has examined in years. Sometimes the finding is that your problem is a process problem and no technology will help. We will tell you that too.
Either way, draw the map. You cannot fix a wait you have never measured. Until it is on paper next to the work, you will keep believing a quote takes forty minutes.
Next Monday: we ran a wastage and gross profit audit on a working kitchen. What the numbers actually showed and why the gap between the spreadsheet and the shelf was so wide.